GitHub Copilot code review can now be requested through REST and GraphQL APIs, with an optional review effort level for each request. In its October 2 announcement, GitHub also confirmed that Balanced had become the default effort level on September 28, 2026, for new and existing repositories and organizations using Copilot code review. Explicit Lite selections were preserved.
The changes are generally available on Copilot Pro, Pro+, Max, Business, and Enterprise plans, according to GitHub. For developers and platform engineers, they introduce two separate decisions: when automation should request a review, and which effort setting that review should use.
API requests bring review into custom workflows
REST and GraphQL support gives teams a programmatic way to request Copilot reviews. A script or internal developer tool can make the request as part of an existing workflow, rather than requiring someone to initiate it manually. The optional effort setting lets the caller select a level for an individual review.
That creates room for workflow-specific choices. A team could decide which review requests should explicitly select an effort level and which should rely on its configured settings. This is an integration opportunity, rather than evidence that a particular workflow will produce better reviews.
Before implementing an integration, teams should check the API documentation for endpoint paths, GraphQL mutations, authentication requirements, and request payloads.
Default now means Balanced
The default change has a different operational consequence: it affects existing configurations as well as new ones. GitHub says the Default setting now uses Balanced for repositories and organizations using Copilot code review. The effective date was September 28, preceding the October 2 API announcement.
An explicit Lite selection remains Lite. Teams therefore need to distinguish between choosing Lite and leaving the setting on Default. A repository that relies on Default now receives the Balanced default, while a previously explicit Lite choice was respected.
The names alone do not establish tradeoffs in review latency, cost, or defect detection between Lite and Balanced. Teams assessing the change should check their settings and evaluate review results against their own needs before drawing conclusions about either level.
Where teams can configure effort
Effort can be configured at enterprise, organization, repository, and personal levels. DiffVibe’s coverage describes each level as overriding the one above it, including repository settings overriding organization settings and organization settings overriding the enterprise default.
For platform teams, the immediate task is to inspect the configuration that applies to their repositories and decide whether Balanced or an explicit Lite setting matches their intended policy. When adding API requests, they should also decide where per-request effort selection belongs in the workflow. That makes both the review trigger and the effort choice deliberate as automated review becomes part of their tooling.
